So! This week follows the trends of recent ones in not having a lot of easy ways to show my progress, so I’ll use words to describe what I’ve done.
Fun with Spreadsheets and Planning
The most major thing I accomplished was the re-planning of tasks I need to complete to test version 0.1.0 of the game. For the past few weeks I’d be using a spreadsheet for this as it allowed me to make dynamic calculations and metric specific predictions that programs like HacknPlan and Trello lacked the flexibility to properly share with my audience.
Here’s what I did before:
Now don’t get me wrong - this isn’t bad exactly - I actually think it’s aesthetically nicer and more readable than what I came up with in the redo. The main problem with it is that it was difficult for me to add and break down unexpectedly complicated tasks into smaller parts. This lead to a problem where the list would become exceedingly out-of-date, and not reflect the hours of extra effort I was applying on a task initially predicted to take only 2 hours.
That was the big drive for me to make the new one, and the new one has two major aspects.
The frontend:
The backend:
So, what are we looking at here and why did I spend multiple days of my week making it? Let’s start with the backend. On the backend you’ll notice that I am indenting tasks by cells. Whenever a task is indented, that means it is considered a sub-task of the unindented one above it. Let’s call that unindented task the parent task, and the tasks organized under it as the child tasks. Parent tasks are purple or yellow, while child tasks are white or green.
Only child tasks are workable tasks, as they are the only ones small enough to have a semi-competent estimate made. The parent tasks are too broad of a concept typically, hence why they’re broken down into more comprehensive ones.
All of this means I now have a more concrete idea of what I need to do for the game, and can easily add or remove things as my understanding improves through making it. The frontend simply takes the information I place in the backend and makes it presentable.
Over time I may improve the frontend, as right now it’s just a very basic list - not super reader friendly. Users who are curious as to what the specific tasks are may have more trouble understanding them. But my main focus is creating that message at the top and keeping it as accurate as possible.
Audio FX
So I originally had a lot of audio FX planned, however I horribly underestimated how long it would take me to get the audio sound right. That means I ended up having to whittle down the audio to the basics:
- footstep sounds for grass, concrete, and metal
- slight wind sound when running quickly
as well as a handful of sounds that got cut because adding the sound mechanic (like armored outfits making clanking sounds when the character moved) would require more audio FX than I have time for. Eventually if I get a larger budget for hiring contractors / more time to do this I might bring these ideas back.
In terms of sharing the audio I’m not sure it’s really impressive enough, and even more I’m not actually sure it’s good enough haha - I’ll want to get some playtesting feedback to verify I did it right before releasing anything. As I said though, even if I do it right it’s not super interesting so by the time I’d be ready to release it I’d hopefully have cooler things to show.
More Game Feel Polish
So I spent a few days this week doing a bunch of miscellaneous tasks like fixing a few weird bugs with the custom physics (like cancelling momentum when you hit a wall so you don’t get stuck there), as well as improvements to the player input translation routines, and even some improvements to the camera system. Once again, these aren’t really super exciting things to show, but I’m happy with how the avatar is turning out. There are still a few areas to polish, especially relating to inverse-kinematics, but I’m fairly confident by the end of next week I’ll be in a pretty good position to move on to the next big step in the journey - secondary assets and blender.
Next Week
So I’ll start the week finishing up on the game feel related stuff by rewriting part of the IK engine to better adjust legs in a way that doesn’t compromise the visuals of an in-progress animation. But that hopefully will be done by Tuesday, allowing me to focus on the main goal: learning Blender.
As some of you may know, I haven’t really used blender before. I’ve worked with Autodesk Maya, older 3D modeling programs for engineers like CATIA, and more casual 3D modeling software like Google Sketchup. I however have not used blender, and frankly it’s starting to hold me back as a developer.
From what I can tell Blender is the best 3D modeling software for indie developers. It’s an accessible, powerful tool, with a strong community + ton of documentation, and no fee for usage on commercial projects. Blender is great, and it will be my best option for creating meshes with videogames.
Due to Roblox characters historically not requiring traditional rigging or even meshes, I was able to avoid learning it for a while. Now that the LP3 uses mostly rigged mesh assets, I can no longer rationalize avoiding it. Currently if I have a problem with a mesh I can ask a contractor, wait a few days, and get back a fix - but it will likely cost be a chunk of money, especially for larger projects. This long term is expensive and pretty slow - and I can fix it by learning Blender.
With the help of my friend Ryferion1, who is responsible for the current amazing LP3 assets, I’ve got myself some blender homework. The goal is to teach myself in around a week or so the bare minimum needed to make what I need to make consistently. It won’t necessarily be the best work, but it will work. Over time I’ll certainly improve, but so long as I can produce assets at the standard I need quickly I will avoid many future gridlocks by bypassing contractors, only using them on items which really do need an artist’s touch, rather than on every little thing.
Estimated Release Date
Currently the estimated release date is November 23rd due to the more realistic predictions made in the new planner. This isn’t so much of an increase in the amount of work, rather a more accurate estimate of how long it will take. We are definitely getting closer to release though, even if it doesn’t feel like it. These game feel improvements have got me really excited to get this prototype out, but I don’t want to sacrifice the reason I’m excited - the quality - for a quicker release. A good game that comes out late isn’t remembered as a late game, while a bad game that comes out on-time is forever remembered as a bad game.
Talk to y’all next week!